Add Encoding.Utf8WithoutBom static extension property

[Copilot]

Encoding.UTF8 emits a BOM, but there's no convenient analog on the Encoding type for the BOM-less variant. This adds a static extension property Utf8WithoutBom backed by new UTF8Encoding(false).

Changes

• PowerKit/Extensions/EncodingExtensions.cs — new extension(Encoding) block exposing Utf8WithoutBom as a static property; instance cached in a private static field on the enclosing class
• PowerKit.Tests/EncodingExtensionsTests.cs — functional test that encodes text using Utf8WithoutBom and verifies it is correctly decoded by Encoding.UTF8
• PowerKit/Extensions/ZipArchiveEntryExtensions.cs — refactored to use Encoding.Utf8WithoutBom for write methods (no BOM emitted) and Encoding.UTF8 for read methods (auto-detects and strips any BOM in the content); entire class wrapped in #if NET40_OR_GREATER || NETSTANDARD || NET since ZipArchiveEntry is not available in .NET 3.5

Usage

using var writer = new StreamWriter(stream, Encoding.Utf8WithoutBom);

[Copilot]

Pull request overview

Adds a convenient BOM-less UTF-8 Encoding via a static extension property, plus tests to validate its behavior, to make it easier to create StreamWriter/text output without an emitted BOM.

Changes:

• Introduces Encoding.Utf8WithoutBom backed by a cached UTF8Encoding(false) instance.
• Adds unit tests asserting UTF-8 identity, empty preamble (no BOM), and non-identity with Encoding.UTF8.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
PowerKit/Extensions/EncodingExtensions.cs	Adds Encoding.Utf8WithoutBom static extension property backed by a cached encoding instance.
PowerKit.Tests/EncodingExtensionsTests.cs	Adds tests covering web name, BOM-less preamble, and distinction from Encoding.UTF8.

Comments suppressed due to low confidence (1)

PowerKit.Tests/EncodingExtensionsTests.cs:26

• Test method names in this repository consistently use the ..._Test suffix (and ..._Condition_Test for variants). These new tests omit the suffix, which makes the file inconsistent with the rest of the test suite; please rename the methods to match the existing naming convention.

    public void Utf8WithoutBom_Test()
    {
        // Arrange
        var text = "hello, world! 🌍";

        // Act
        var bytes = Encoding.Utf8WithoutBom.GetBytes(text);

        // Assert
        Encoding.UTF8.GetString(bytes).Should().Be(text);
    }
}

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Tyrrrz]

@copilot revert last change, there is no Encoding.ReadOnly(...)

[Copilot]

@copilot revert last change, there is no Encoding.ReadOnly(...)

Reverted in 6633d98 — back to new UTF8Encoding(false) directly.

[Tyrrrz]

@copilot merge in latest prime and see if you can refactor some existing code with the newly added property

[Copilot]

@copilot merge in latest prime and see if you can refactor some existing code with the newly added property

Done in 0f4e329. Merged prime and refactored ZipArchiveEntryExtensions — it had its own private Utf8NoBom = new UTF8Encoding(false) field that's now replaced with Encoding.Utf8WithoutBom across all 8 usages. All 133 tests pass.

[Tyrrrz]

@copilot

1. Build failing.
2. Do you think the read methods in ZipArchiveEntry extensions should actually use normal Encoding.UTF8 so that they can also handle conent with BOM?

[Copilot]

@copilot

1. Build failing.
   ...

Fixed in 97c15d9.

1. Build fix: The entire ZipArchiveEntryExtensions class is now wrapped in #if NET40_OR_GREATER || NETSTANDARD || NET since ZipArchiveEntry / System.IO.Compression don't exist in net35. The inner #if that previously only guarded the async methods was redundant and is removed.

2. Read vs write encoding: Yes — read methods now default to Encoding.UTF8, which causes StreamReader to auto-detect and strip any BOM present in the content. Write methods keep Encoding.Utf8WithoutBom so no BOM is ever emitted.